Endoscope Spatial Information Correction Using Reference Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing endoscope systems face challenges in accurately acquiring and processing spatial information of lumens, leading to incomplete observations and estimation errors due to the non-rigid nature of organs and accumulation of calculation errors during depth map creation.
Innovation Solution
A medical information processing apparatus and method that sequentially acquires spatial information from endoscopic images, estimates a three-dimensional environment map or distal end position of the endoscope, and corrects these using reference information about the shape and absolute size of the lumen, ensuring accurate medical information acquisition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If depth map is created using sequential endoscopic images, then observation completeness can be checked, but estimation errors accumulate and measurement precision deteriorates
Solution Approach 1:
The system uses pre-acquired reference information (from CT, MRI, or ultrasound) as feedback to correct the depth map and spatial information during endoscopic examination. This feedback mechanism compensates for accumulation errors in real-time depth estimation, maintaining measurement precision while enabling complete observation through sequential imaging.
Solution Approach 2:
Reference information about the lumen's three-dimensional shape is acquired before the endoscopic examination (preliminarily). This pre-acquired data serves as a foundation for correcting subsequent real-time depth map calculations, preventing error accumulation before it occurs during the actual observation process.
2Measurement precision
If reference information from CT or MRI is used to correct depth map, then measurement precision improves, but device complexity increases
Solution Approach 1:
The medical information processing apparatus is designed to accept and process multiple types of reference information (CT, MRI, ultrasound) and integrate them with endoscopic images. This multi-functional capability allows the system to work with various imaging modalities without requiring separate dedicated systems, managing complexity through universal processing architecture.
3Loss of information
If three-dimensional environment map is created from sequential images, then diagnostic information quality improves, but calculation time and processing complexity increase
Solution Approach 1:
Reference three-dimensional information is acquired and prepared before the endoscopic examination. This preliminary preparation allows the system to perform rapid corrections during the actual procedure by comparing real-time depth estimates against pre-computed reference data, reducing real-time calculation time while maintaining high diagnostic information quality.
Data Source
AI summary
A medical information processing apparatus according to an aspect of the present invention is a medical information processing apparatus including a processor. The processor is configured to execute a spatial information acquisition process of sequentially acquiring pieces of spatial information of a lumen on the basis of endoscopic images sequentially acquired by an endoscope; an estimation process of estimating, on the basis of the pieces of spatial information sequentially acquired, at least one of a three-dimensional environment map of the lumen or a distal end position of the endoscope; a reference information acquisition process of acquiring reference information about a shape and/or an absolute size of the lumen; and a correction process of correcting, using the reference information, one or more of the spatial information, the three-dimensional environment map, and the distal end position of the endoscope.


